Crack Detection on Surfaces Using Digital Image Processing

The extraction and connecting of break sites of cracks in pavement photographs is presented in this study using a unique, rapid, and self-adaptive image processing approach. This method was stimulated using matlab. The method determines the crack's categorization into transverse and longitudinal kinds by first locating the crack's start location. For various types of fractures, different search techniques are utilized. The program then traces along the crack pixels to discover the break site and connects it to the nearest crack point in the search section. The closest point in the region becomes the new start point, and the algorithm repeats the procedure till it reaches the crack's conclusion. This experiment was carried out on a number of pipes amd result was accuarate for most pipes. The results of the experiments reveal the connection approach is quite successful in maximizing crack identification accuracy.